Datasets and Models
Download Links:
Everything: osf.io/tuwh5
CT Dataset: Leg
CT Dataset: Fullbody
CT Dataset: Lowerbody
Synthetic Dataset
Code
Download Links:
Everything: Github
Unity Renderer
Gaussian Splatting Training
Path Tracer (Dataset Generation)
In medical image visualization, path tracing of volumetric medical data like CT scans produces lifelike three-dimensional visualizations. Immersive VR displays can further enhance the understanding of complex anatomies. Going beyond the diagnostic quality of traditional 2D slices, they enable interactive 3D evaluation of anatomies, supporting medical education and planning. Rendering high-quality visualizations in real-time, however, is computationally intensive and impractical for compute-constrained devices like mobile headsets.
We propose a novel approach utilizing GS to create an efficient but static intermediate representation of CT scans. We introduce a layered GS representation, incrementally including different anatomical structures while minimizing overlap and extending the GS training to remove inactive Gaussians. We further compress the created model with clustering across layers.
Our approach achieves interactive frame rates while preserving anatomical structures, with quality adjustable to the target hardware. Compared to standard GS, our representation retains some of the explorative qualities initially enabled by immersive path tracing. Selective activation and clipping of layers are possible at rendering time, adding a degree of interactivity to otherwise static GS models. This could enable scenarios where high computational demands would otherwise prohibit using path-traced medical volumes
Fullbody Layers
Synthetic 3Layers
Note how layers build upon previous ones. Download the CT and synthetic datasets from osf.io/tuwh5
@misc{kleinbeck_multi-layer_2024, title = {Multi-{Layer} {Gaussian} {Splatting} for {Immersive} {Anatomy} {Visualization}}, url = {http://arxiv.org/abs/2410.16978}, doi = {10.48550/arXiv.2410.16978}, publisher = {arXiv}, author = {Kleinbeck, Constantin and Schieber, Hannah and Engel, Klaus and Gutjahr, Ralf and Roth, Daniel}, month = oct, year = {2024}, note = {arXiv:2410.16978}, keywords = {Computer Science - Computer Vision and Pattern Recognition, Computer Science - Graphics}, }
We gratefully acknowledge funding for this study by d.hip campus.